Skip to content

Meeting room stickman

image

Background

Using a mounted camera with facial recognition one should be able to count the number of people in a meeting room.


Task

The main task of this project is to develop a room display system that automatically can see how many (and possibly who) is attending a meeting. The system should also identify where in the room the person are sitting and display this on a map of the room.

Main task:

  • Develop a system that maps the camera's view to the physical layout of the room and the persons within.
  • Detect the presence and position of all persons in the room and their positions.
  • Send this information to a server where it will be translated to a image to display on the meeting room display system.

Bonus task:

  • Determine who is in the meeting. The persons are already known to the system.

Epic

This system will build upon prior work about facial recognition light-controller project, which provides a basic foundation.


Legal and Technical Requirements

  • All source code must be released under a copyleft license (e.g., GPLv3).
  • Emphasis should be on product readiness, stability, and maintainability.
  • Code must follow best practices, standardized language conventions, and provided templates.
  • Automated checks, tests, and build procedures (via GitLab CI) should be implemented where possible.
  • The final deliverable (build package) must be directly installable on the host operating system.
Edited by Jonas Karlsson
To upload designs, you'll need to enable LFS and have an admin enable hashed storage. More information